Railway-mcp: Instructions file for Claude Code

CLAUDE.md

Railway-mcp CLAUDE.md is an instructions file for Claude Code from Travis-Gilbert/Railway-mcp. It costs 837 tokens per session, scanned A, original, MIT.

Project instructions for a Python server that connects Claude Code to Railway's GraphQL API. GraphQL is a way for software to request specific data from a service.

In plain words
What is it for?
Use them when developing, debugging, or navigating the Railway MCP server and its API-related files.
Why use it?
They give the agent the project's entry points, file layout, authentication setup, and patterns for adding or maintaining tools.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md.

This is Travis-Gilbert/Railway-mcp's own configuration. It tells Claude Code how to work on Railway-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Railway-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Travis-Gilbert/Railway-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Travis-Gilbert/Railway-mcp/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Travis-Gilbert/Railway-mcp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for Railway-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/travis-gilbert/railway-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/travis-gilbert/railway-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/travis-gilbert/railway-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/travis-gilbert/railway-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for Railway-mcp CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/travis-gilbert/railway-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/travis-gilbert/railway-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 837 This file is loaded in full into every session.
When invoked 837 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00837 $0.00837
Opus 5 $0.00418 $0.00418
Sonnet 5 $0.00167 $0.00167
Haiku 4.5 $0.00084 $0.00084

Measured 9d ago against content hash 90b860c8f1b6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

Railway-mcp CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Railway MCP Server

Python FastMCP server wrapping the Railway GraphQL API for use in claude.ai.

Quick Reference

  • Entrypoint: python -m railway_mcp
  • All tools prefixed with railway_
  • GraphQL endpoint: https://backboard.railway.app/graphql/v2
  • Auth: Bearer token via RAILWAY_API_TOKEN env var
  • Transport: streamable-http on Railway, stdio for local dev

File Map

  • railway_mcp/server.py - FastMCP instance + all 23 tool registrations
  • railway_mcp/client.py - Async GraphQL client with error handling
  • railway_mcp/queries.py - All GraphQL query/mutation string constants
  • railway_mcp/models.py - Pydantic v2 input models for every tool
  • railway_mcp/formatting.py - Markdown/JSON response formatting helpers
  • railway_mcp/__main__.py - Entrypoint (transport + port from env)
  • railway_mcp/__init__.py - Package marker

Architecture

This follows the exact same pattern as Travis-Gilbert/ticktick-mcp:

  • FastMCP handles MCP protocol + transport (streamable-http or stdio)
  • httpx.AsyncClient talks to Railway's GraphQL v2 API
  • Pydantic models validate all tool inputs
  • Every tool returns formatted markdown or JSON based on response_format param

Key Patterns

  • The GraphQL client is a module-level singleton (get_client())
  • All tools catch RailwayAPIError and return error strings (never raise)
  • Relay-style pagination (edges/node) is extracted via _extract_edges()
  • Destructive tools (delete_variable, bulk_set with replace=True, delete_service) have destructiveHint=True
  • Projects/services/environments are referenced by UUID strings
  • Railway's staged config model: serviceInstanceUpdate writes config, then serviceInstanceDeployV2 applies it. The tools guide users through this two-step flow via descriptions.

Tool Count

23 total: 10 read-only, 13 write (3 destructive)

Categories:

  • Projects (2): list, get
  • Services (6): list, get, create, delete, connect, disconnect
  • Service Config (1): update (Dockerfile path, start command, root dir, health check, replicas)
  • Environments (3): list, create, duplicate
  • Variables (5): list, list unresolved, set, bulk set, delete
  • Deployments (6): status, build logs, deploy logs, redeploy, deploy (fresh), restart

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 89 lines · 837 tokens per session scan A 90b860c8f1b6

Subscribe to this mod's changes

Railway-mcp CLAUDE.md is an instructions file published in the GitHub repository Travis-Gilbert/Railway-mcp (0 stars, last pushed 5mo ago), licensed MIT. It adds 837 tokens to every session, about $0.0042 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories